UPVC Class 16 and PN 16 — When You Actually Need It

The Question This Page Answers

You have a UPVC line to specify and someone has written “Class 16” or “PN 16” on the drawing. The question is not what those designations mean — our pipe sizing and pressure class reference already sets out how PN is defined and how SDR converts to a rating. The question here is narrower and more expensive to get wrong: does this service actually need 16 bar of UPVC, or is 16 bar the number someone reached for because it sounded safe?

In our experience quoting UPVC pressure lines, the same three situations keep appearing. One: the job genuinely needs 16 bar — a pump-fed riser, a long transmission run, real surge exposure. Two: the job needs 6 to 10 bar and Class 16 was written in for margin, buying wall thickness that does nothing except raise cost and shrink bore. Three, the dangerous one: the job is written as Class 16 when the actual problem is temperature, and no pressure class solves a temperature problem. That third case is where UPVC gets blamed for failures that were specification failures.

First, the Trap: “Class 16” Is Five Different Things

Before any application decision, you need to know which document the buyer is quoting, because the same phrase resolves to different physical pipe — or to nothing at all — depending on origin. This is specifically about the word Class, which is messier than PN.

Of five recognised UPVC pressure standards, exactly one issues a “Class 16”: the South African SANS 966-1. Everywhere else the phrase is either an informal translation of PN 16 or — in North America — a unit error, because ASTM Class numbers are PSI. So if you are tendering into South Africa or a market that took SANS as its reference, honour “Class 16” as written. Anywhere else, ask which standard applies rather than assuming PN 16 and shipping; in our experience that one question resolves more UPVC specification disputes than any technical argument that follows it.

Sourcing note: the SANS 966-1 and AS/NZS 1477 values were verified from manufacturer catalogues and data sheets citing those standards, not from the paywalled standard texts. Treat them as reliable manufacturer restatements rather than quoted clauses.

Class 16 and PN 16: Same Pressure, Same Design Stress, Different Rounding

Both “Class 16” (SANS) and “PN 16” (ISO) mean sixteen bar. The published wall thickness columns agree exactly up to dn 90 and then differ by 0,1 to 0,3 mm — a difference worth understanding before it turns into an argument at goods inwards, because it is smaller and more mundane than it looks. Comparing the SANS 966-1 Class 16 minimum wall thickness column against ISO 1452-2:2009 Table 2 for PN 16:

Sources: SANS Class 16 column from the Swan Plastics AQUAFLO catalogue to SANS 966 Part 1:2014; ISO PN 16 column from ISO 1452-2:2009(E) Table 2 — a direct numerical read of the two published tables.

The cause is a design-coefficient split written into ISO 1452-2:2009 Table 2 itself: PN values for dn 12–90 mm are “based on design coefficient C = 2,5”, while PN values for dn 110–1000 mm are “based on design coefficient C = 2,0” (ISO 1452-2:2009(E), Table 2). Since the material must have a minimum required strength (MRS) of at least 25 MPa (ISO 1452-2:2009(E), Clause 4.3), design stress works out at σs = 25/2,5 = 10 MPa below 90 mm and 25/2,0 = 12,5 MPa from 110 mm up. Higher permitted design stress means a thinner wall for the same 16 bar.

That explains the step at dn 110 in the ISO column. It does not explain the gap between the two columns, and it is worth being precise about what the gap actually is. Reversing the standard wall equation e = P·dn / (2σs + P) on each printed value — as σs = P·(dn − e) / 2e, with P = 1,6 MPa — returns the design stress each column was built on:

• dn 50 to 90, both columns: σs comes back at 9,9 to 10,0 MPa — the C = 2,5 case, as expected.
• dn 110 to 315, both columns: σs comes back at 12,3 to 12,7 MPa — the C = 2,0 case, in the SANS column just as much as the ISO one.

So the two tables are not built on different design stresses. Both sit on σs = 12,5 MPa above dn 90. The reason the reversed values scatter either side of 12,5 rather than landing on it is that the published walls are rounded to 0,1 mm, and reversing the equation amplifies that rounding. Work the formula forwards at σs = 12,5 MPa and the position is unambiguous: dn 110 gives e = 1,6 × 110 / (2 × 12,5 + 1,6) = 176 / 26,6 = 6,617 mm; dn 160 gives 256 / 26,6 = 9,624 mm; dn 315 gives 504 / 26,6 = 18,947 mm. The SANS column is exactly those figures rounded up to the next 0,1 mm — 6,7 / 9,7 / 19,0, matching the printed SANS value at every one of the six diameters above dn 90 in the table. The ISO column sits just below the same computed figures (printed 6,6 / 9,5 / 18,7 against computed 6,617 / 9,624 / 18,947); we have not established which rounding or series-derivation rule ISO applies to get there, and we are not going to guess one. What the arithmetic does establish is the part that matters commercially: both columns are generated from the same σs = 12,5 MPa above dn 90, so the difference between them is a rounding and derivation artefact of the order of 0,1 mm — not a deliberate difference in engineering conservatism between the two standards.

Three things follow that matter on a purchase order:

• Below 110 mm the distinction is academic. Same wall either way. If your scope is 20–90 mm, stop worrying about which word the drawing used.
• From 110 mm upward the two printed columns differ by 0,1 to 0,3 mm, and that difference is a rounding artefact on a shared σs = 12,5 MPa, not a different design basis. Both are 16 bar pipe. Do not let the two columns be read as two grades of pipe, and do not let a 0,1 mm delta on paper become a conformance argument on its own — conformance is judged against the wall thickness tolerance the governing standard specifies, which is a separate question from the difference between two nominal columns. If your contract names one standard, take the wall column from that standard and settle acceptance on its tolerance clause.
• ISO 1452-2:2009 Table 2 footnote a gives the formal escape route: to apply C = 2,5 rather than 2,0 above dn 90, “the next higher PN shall be chosen”. So a buyer wanting the conservative coefficient on large-diameter ISO pipe specifies PN 20, not PN 16 — the clause to cite when someone asks you to “make PN 16 a bit thicker”.

For our own scope this is largely theoretical: our UPVC 806 line runs Φ20 to Φ110 in nine sizes, 4 m lengths (per our catalogue — full wall table further down this page), so only the largest size reaches dn 110, the point where the two columns stop being identical. Below that, every one of our sizes sits in the range where SANS Class 16 and ISO PN 16 print the same wall. We do not manufacture PVC-U pressure pipe above Φ110 at all.

Where 16 Bar Is Genuinely Justified

These are the conditions where, in our experience selling UPVC into distribution and contracting, 16 bar is a defensible decision rather than a habit. None are cited to a standard — they are judgement calls, labelled as such.

• Pump discharge and shut-head exposure. What drives the class is not running pressure but what the pipe sees when something goes wrong. A booster set running at 5 bar can deadhead well above that, and the pipe between the pump and the first relief device sees it. Where shut-off head plus static lift approaches 10 bar, 16 bar is sensible; where a properly sized relief valve caps the system well below that, it is not.
• Tall risers where static head alone is significant. Static head never gets designed out — roughly one bar per ten metres of column before any pump contribution. That is arithmetic, not a standard, but it regularly pushes the bottom zones of a tall building into a genuine 16 bar requirement while upper zones sit comfortably at PN 10.
• Long buried runs with surge exposure. A fast-acting valve or unprotected pump trip generates transients well above operating pressure, and UPVC is comparatively rigid — it does not accommodate them by deforming. With no surge vessel, soft-start or slow-closing valves, treating 16 bar as the class is reasonable. Quantifying the surge is specific to your run; we publish no generic surge allowance because there is no defensible generic figure.
• Where the specification is a compliance obligation. Sometimes 16 bar is required because a water authority or tender document says so — a legitimate reason to buy Class 16 even when the hydraulics do not demand it. But it is a procurement requirement and should not be presented to the client as a technical necessity.

Where 16 Bar Is Over-Specification

The counterpart list — again our observations, not standard clauses.

• Gravity-fed and tank-fed distribution below three storeys. Maximum static head under three bar with no pump in the circuit means a 16 bar class buys five times the pressure the system can physically generate. PN 10 or even PN 6 covers it with margin.
• Irrigation laterals downstream of a pressure regulator. The regulator defines maximum downstream pressure. Laterals at 16 bar behind a regulator set at 3 bar is a mismatch — spend the money on regulator reliability instead.
• Branch runs off a properly zoned riser. If the riser is zoned with pressure-reducing valves, branches see the reduced zone pressure. Specifying every pipe on the job to the worst-case riser class is common and almost always wasteful.
• Any non-pressure service. Soil, waste and rainwater lines have no meaningful pressure class. Our PVC 902 range is non-pressure drainage (per our catalogue) — 902 pipe in Φ32 to Φ110, with 1902 fittings up to Φ160 — not a Class 16 product, not a low-pressure alternative to one, and never a substitute in a pressure line.

There is also a bore penalty that rarely gets costed. A heavier class at the same outside diameter means a thicker wall, a smaller bore, and higher friction loss at the same flow — on a long run that can quietly push you up a nominal size, so the “safe” choice costs more twice over. We have no measured head-loss figures for our own line to publish: Coming soon.

The Decision Table

A condensed version of the two sections above, for use when the drawing is ambiguous. The right-hand column is our view, offered as engineering judgement rather than as a standard requirement.

The Real Failure Mode: Buying Pressure Class When the Problem Is Temperature

Pressure class and temperature capability are independent properties. Going from PN 10 to PN 16 buys wall thickness against pressure. It buys effectively nothing against heat, because derating applies to whatever class you started from. A PN 16 pipe at elevated temperature is a derated PN 16 pipe — the class does not exempt it.

Two hard limits from the standards, both more restrictive than the figures that circulate informally:

• PN is declared for water up to and including 25 °C — not 20 °C. Clause 1 states the standard “is applicable to piping systems intended for the supply of water under pressure up to and including 25 °C (cold water)” (ISO 1452-2:2009(E), Clause 1; Annex A.2 title). The widely repeated “PN is at 20 °C” is the PE convention, not the PVC-U one.
• PVC-U is capped at 45 °C, not 60 °C. Clause 1: pipes “for the conveyance of water and waste water up to and including 45 °C. For temperatures between 25 °C and 45 °C, Figure A.1 applies.” NOTE 1 adds that use above 45 °C is only by case-by-case agreement between producer and end-user; Annex A is normative (ISO 1452-2:2009(E)).

The derating is steep: the coefficient on allowable operating pressure is 0,80 at 35 °C and 0,63 at 45 °C, referenced to 25 °C (M/TEC-1452 datasheet to UNI EN ISO 1452 — a manufacturer restatement, not the ISO text). Applied to a PN 16 line:

We deliberately do not fill in 30 °C or 40 °C. The complete ISO 1452-2 Figure A.1 curve is behind the paywall — the free ISO sample truncates before Annex A — and the two anchor points come from a manufacturer datasheet citing the standard, one version of which is internally inconsistent between its language columns on which temperature corresponds to 0,63. Interpolating on that basis would be inventing engineering data: intermediate values, Coming soon.

Read that against the decision problem. At 45 °C a PN 16 UPVC line delivers about 10 bar — roughly what a PN 10 line delivers cold. If your specification jumped to Class 16 to cope with a warm service, you paid for wall thickness the temperature has already spent.

The North American figures are on a different basis

Projects mixing European and North American references produce two irreconcilable sets of temperature figures for nominally the same polymer. Both are correctly sourced; the bases differ. The ASTM-derived system references ratings to 73 °F / 23 °C and permits PVC to 140 °F / 60 °C, above which it is listed “NA”, with the ladder: 73 °F = 1.00, 80 °F (27 °C) = 0.88, 90 °F (32 °C) = 0.75, 100 °F (38 °C) = 0.62, 110 °F (43 °C) = 0.51, 120 °F (49 °C) = 0.40, 130 °F (54 °C) = 0.31, 140 °F (60 °C) = 0.22, the method being “multiply 73 °F (23 °C) pressure rating by de-rating factor” (Spears Manufacturing, De-Rating table, Page 17). The PVC Pipe Association independently gives the same 60 °C maximum and 23 °C reference, with a factor of 0.20 at the maximum (Westlake technical FAQ, citing Handbook of PVC Pipe Design and Construction, 5th Ed.).

At 100 °F — 38 °C, unremarkable for a rooftop run in the Gulf or West Africa — the factor is 0.62, so a 16 bar cold rating becomes roughly 10 bar; at 49 °C it is 0.40, about 6,4 bar. Neither is a defect; both are published material behaviour. But a buyer who specified Class 16 for margin and then ran the line unshaded on a roof has consumed all of it. So: state the design temperature on the enquiry. A pressure class without a temperature is an incomplete specification.

When You Should Leave PVC-U Entirely

If the service is genuinely hot, the answer is a different standard family, not a heavier class.

PVC-U pressure pipe is governed by ISO 1452. Chlorinated PVC is a separate family: ISO 15877, “Plastics piping systems for hot and cold water installations — Chlorinated poly(vinyl chloride) (PVC-C)”, scoping pipes for “hot and cold water installations within buildings … and for heating systems, under design pressures and temperatures appropriate to the class of application”, with application classes defined by design temperature TD, maximum Tmax and malfunction Tmal (ISO 15877-2:2009(E), Clause 1 Scope).

The standard-level evidence of the gap is in qualification testing: Clause 4.3 requires PVC-C to be qualified by long-term hydrostatic testing at “20 °C; 60 °C to 70 °C; 95 °C” for Type I and “20 °C; 60 °C to 70 °C; 100 °C” for Type II (ISO 15877-2:2009(E)). Those are material qualification temperatures, not permitted continuous service temperatures — a distinction frequently blurred. We could not verify the TD / Tmax / Tmal values per application class from ISO 15877-1 Table 1, so we do not quote them: Coming soon.

On derating, the same Spears table gives CPVC: 73 °F = 1.00, 90 °F (32 °C) = 0.91, 100 °F (38 °C) = 0.82, 120 °F (49 °C) = 0.65, 140 °F (60 °C) = 0.50, 160 °F (71 °C) = 0.40, 180 °F (82 °C) = 0.25, 200 °F (93 °C) = 0.20 (Spears Manufacturing, PVC & CPVC De-Rating table, Page 17). At 140 °F CPVC retains 0.50 against PVC’s 0.22, and continues to 200 °F where PVC is listed as unusable. The physical reason is chlorine content raising the temperature at which the polymer softens: a CPVC producer publishes HDT (ASTM D648) = 217 °F (103 °C) and maximum pressurised service 200 °F (93 °C) (Corzan CPVC / Lubrizol) — producer-branded values for a specific compound, not standard-mandated minima, so read them as illustrative.

For PVC-U’s own material envelope, independent European manufacturer data gives Vicat softening point (B/50, ISO 306) = 76 °C and HDT (ASTM D648, 0,46 N/mm²) = 86 °C, with a stated service range of 0–60 °C referred to the pressure/temperature regression curves, MRS = 25 N/mm², classification PVC-U 250 (FIP PVC-U Technical Handbook). Note the distinction catalogue claims routinely collapse: a material envelope of 0–60 °C is not the pressure application limit of 45 °C in ISO 1452-2. A pipe can survive a temperature it cannot be pressurised at.

Our UPVC 806 Line, Stated Honestly Against All of the Above

Everything in this section is from our catalogue and nothing in it is derived, extrapolated or calculated.

The full 806 wall thickness table

Nine sizes, every row as printed in our catalogue. Nothing here is interpolated or reverse-calculated from the design equation — read it against the ISO and SANS columns earlier on this page if you are cross-checking.

All nine at PN16, 4 m lengths. Note that Φ50×3,7, Φ63×4,7, Φ75×5,6 and Φ90×6,7 land exactly on both the SANS 966-1 Class 16 and the ISO 1452-2 PN 16 columns given earlier — which is what you would expect below dn 110, where the two standards agree. At Φ110 our 7,2 mm sits above both the SANS 6,7 and the ISO 6,6, so the top of our range is thicker than either 16 bar column requires, not thinner.

Three points of honesty a specifier deserves before enquiring:

On the 95–120 °C figure. That is what our catalogue states as heat resistance for the UPVC/CPVC 806 system, quoted verbatim — the heat-resistance wording in the catalogue is written for the CPVC grade, and we do not restate it as a UPVC-only figure. It is a material heat-resistance claim, not permission to operate a PN16 UPVC line at 95 °C under pressure — ISO 1452-2:2009 scopes PVC-U pressure application to 45 °C, above which it is case-by-case agreement between producer and end-user (ISO 1452-2:2009(E), Clause 1 NOTE 1). If your service is hot and pressurised, do not read our catalogue heading as a green light; ask us in writing.

On diameter. Our pressure UPVC stops at Φ110. For DN 150 and above we cannot supply, and will not imply otherwise. In our PVC 902 range the 1902 fittings reach Φ160, while 902 pipe stops at Φ110 — and the whole range is non-pressure drainage (per our catalogue), not a substitute for a Φ160 pressure pipe. Anyone offering a drainage-series pipe against a pressure-class line item is either confused or hoping you are. Our other lines do not rescue the large-diameter case either: PPR PN20 is made only in 20×2,8 / 25×3,5 / 32×4,4 mm, and HDPE PN16 runs nine sizes — Φ20×2,3 / Φ25×2,3 / Φ32×3,0 / Φ40×3,7 / Φ50×4,6 / Φ63×5,8 / Φ75×6,8 / Φ90×8,2 / Φ110×10,0 — and also stops at Φ110 (both per our catalogue). Φ110 is the ceiling for pressure pipe across our whole range; the only thing that reaches Φ160 is the 1902 drainage fitting series.

On origin. IFANNova is a French brand; manufacturing is by Zhuji Fengfan Piping in Zhuji, Zhejiang, China. Nothing in our range is made in France. We state this because a “Class 16 to European standard” tender line sometimes carries an unspoken assumption about origin, best corrected before an order rather than after.

Frequently Asked Questions

Is Class 16 the same as PN 16? In pressure terms yes — both mean sixteen bar. The published wall columns match exactly to dn 90 mm. From dn 110 upward they differ by 0,1 to 0,3 mm, but reversing the wall equation shows both columns are built on the same design stress of 12,5 MPa there, so that gap is a rounding and derivation difference rather than a different design basis. Separately, ISO itself steps from design coefficient C = 2,5 below dn 110 to C = 2,0 from dn 110 up, which is why the ISO wall does not scale smoothly across dn 90 to 110 (ISO 1452-2:2009(E), Table 2).

Is there such a thing as ASTM Class 16? No. In the North American system “Class” is a PSI figure — Class 100, 125, 160, 200, 315 at 73 °F (Spears, Dimensions & Pressure Ratings). A specification calling for “Class 16 to ASTM” needs correcting at source.

Can I run 16 bar UPVC at 60 °C? Not under ISO 1452-2:2009, which scopes PVC-U pressure pipe to 45 °C, above which it is case-by-case agreement between producer and end-user (ISO 1452-2:2009(E), Clause 1 NOTE 1). On the North American basis PVC is listed to 140 °F / 60 °C but with a factor of only 0.22 there (Spears, p.17). Either way, 16 bar is not what you get.

Do you publish wall thicknesses for the sizes between Φ20 and Φ110? Yes — all nine rows are in the table above: Φ20×2,0 / Φ25×2,0 / Φ32×2,4 / Φ40×3,0 / Φ50×3,7 / Φ63×4,7 / Φ75×5,6 / Φ90×6,7 / Φ110×7,2, all PN16 in 4 m lengths. Those are catalogue figures, not values reverse-calculated from the design equation. What we do not publish is anything above Φ110 in pressure UPVC, because we do not make it.

Where to Go Next

• Pipe sizing charts and pressure class reference — the definitions layer behind this page: DN, NPS, SDR and schedule, and how PN is defined in the flange world versus the plastics world.
• UPVC and CPVC pipe and fittings — the 806 PN16 system, nine sizes Φ20 to Φ110 with full wall thicknesses, and the 1806 fittings series (203 items, Φ20–Φ110, per our catalogue).
• HDPE pipe and fittings — PN16, nine sizes Φ20 to Φ110 with full wall thicknesses, with the 603/604 compression series (per our catalogue).
• PPR pipe and fittings — PN20 in 20 / 25 / 32 mm only (per our catalogue).
• PVC 902 drainage — 902 pipe Φ32–Φ110, 1902 fittings to Φ160, non-pressure drainage only (per our catalogue).
• Certifications — SKZ, CE, WRAS, DVGW, SGS, ISO 9001, ISO 14001; certificate numbers Coming soon.
• For importers and distributors — working with buyers specifying to ISO, SANS or ASTM references.

Get the Class Checked Against Your Actual Conditions

Send us four things and we will tell you honestly whether Class 16 is warranted, whether a lower class is sufficient, or whether the service belongs outside UPVC altogether: the maximum system pressure including pump shut-off and static head, the design temperature including any exposed or rooftop runs, the largest diameter in the scope, and the standard your inspector will check against at goods inwards.

If the answer is that we cannot supply it — above Φ110, or a hot pressurised service — we will say so rather than quote around it. Contact IFANNova.

IFANNova is a French brand; manufacturing is by Zhuji Fengfan Piping, Zhuji, Zhejiang, China. Nothing in our range is made in France.